Adolescents ’ maps about well-being , distress and self-destructive trajectories : What ’ s in their voices ?

The mapping of adaptive and maladaptive trajectories in adolescence has been a key concern of developmental psychopathology research. Given the importance of studying adolescents’ own expert views of their experiences, we explored the factors that contribute to the adoption of trajectories characterized by well-being, by distress, and self-destructive trajectories, in a convenience sample of 33 community adolescents (13-21 years old) organized in five focus groups. Adolescents’ mapping of their own trajectories emerged through textual data analysis and was composed mainly of family and individual factors. We proposed a systemic hypothesis to explain how the interactivity between family and individual factors may foster different trajectories: family relational climate and parental emotional support create a context where self-regulation and positive self-esteem are increased. This study contributes to the deeper understanding of developmental trajectories and enriches reflections on the conceptualization and implementation of preventive and therapeutic interventions.


INTRODUCTION
Although most teenagers follow a developmental trajectory that is characterised by a relative lack of emotional or relational difficulties (Loh & Wragg, 2004) and relatively high levels of psychological health, the study of risky and maladaptive trajectories has been a key concern for developmental psychopathology research. The longer a maladaptive trajectory persists, the more difficult it is to return to an adaptive one (Soares, 2000). Negative adolescent trajectories, including self-destructive thoughts and behaviours (SDTB), deserve special attention from clinical and basic science research, as they are one of the primary causes of death in adolescents and young adults (15 to 24 years), and they represent a severe public health problem (WHO, 2006). However, the majority of studies on this issue have included only clinical samples, and only recently have community samples of adolescents been studied, mainly using quantitative methodologies (Cheng et al., 2009;Prinstein, 2008). There is a notable lack of qualitative studies in the literature devoted to understand the meanings, beliefs, judgments, attributions and the processes embedded in these maladaptive trajectories, especially those held by parents, teachers and community adolescents. Given the obvious value of learning about adolescents' views as perceived experts on their own experiences (Mitchell, Kuczynski, Tubbs, & Ross, 2010), we chose to hear 'adolescents' voices', searching their own socio-cognitive 'maps' for factors associated with developmentally adaptive trajectories characterized by well-being, risk trajectories marked by distress, or maladaptive trajectories characterized by SDTB. Accordingly, we contributed to the deeper knowledge of the risk and protective factors in adolescence. We adopted the perspective of Developmental Psychopathology, assuming that different trajectories -along a continuum from adaptive to maladaptive -are inf luenced by (and also actively inf luence) the individual and the context and that may have a multiplicity of causes and maintaining factors (Cicchetti & Rogosch, 2002).

Adaptive trajectories and well-being
Adolescent well-being can be conceptualized through two different, though associated, constructs. Subjective well-being focuses on the selfevaluation of happiness, life satisfaction and positive emotional experiences. Psychological well-being focuses on mental health (Durán, Hernández, & Morales, 2011), assuming that well-being must be defined through the notion of positive psychological functioning based on the presence of adequate resources from different systemic levels -individual, relational and social -, all of which may contribute to adequate trajectories throughout the life cycle (Cebulla, 2009).
Youngsters who report higher levels of well-being, self-esteem, short-term and long-term projects and goals, as well as personality characteristics of f lexibility and optimism, socio-emotional skills (e.g., high tolerance to frustration, self-control, assertiveness, conf lict resolution) and effective coping strategies, are at lower risk levels for psychological suffering (Loh & Wragg, 2004).
The capacity of the family system to adapt to different dynamics imposed by the stage of adolescence (e.g., negotiation, monitorization, the ability to balance parental control and the need for autonomy in the youngsters, the stimulation of routines and familiar rituals) has been consistently found to promote well-being and to protect against maladaptive trajectories (Randell, Wang, Herting, & Eggert, 2006;Van Renen & Wild, 2008). The family sys-tem catalyses the development of social skills, represents manifestations of emotional support and contributes to a sense of the adolescent's own value within both the family and society (Crespo, Kielpikowski, Pryor, & Jose, 2011).
The literature has also consistently revealed the impact of several family variables on adolescents' well-being. These variables include: the dynamics of communication and time shared, positive parenting styles and secure patterns of attachment (Cruz, Narciso, Muñoz, Pereira, & Sampaio, 2013;Jurich, 2008;Tuval-Mashiach, Walsh, Harel, & Shulman, 2008). In particular, family cohesion seems to be strongly associated with adolescent security, allowing adolescents to explore new contexts, and seems to be a protective factor that directly prevents negative behaviours and symptom manifestations. Marital subsystems also play a protective role, as a positive marital relationship decreases the likelihood of risky behaviours in adolescents (Abbot-Chapman, Denholm, & Wyld, 2008). Furthermore, the quality of family relationships and functioning, i.e., the relational climate, has a strong impact not only on adolescent lifeincreasing pro-social behaviours, emotion regulation, self-esteem and positive attitudes towards adolescents' own lives (Turtiainen, Karvonen, & Rahnkonen, 2007) -but it also seems to have positive long-term effects throughout life (e.g., romantic attachment patterns, life goals, skills to deal successfully with life events; Crespo et al., 2011).
School and peers also represent significant systems in adolescents' lives. Integration into peer groups, most of them developed in the school system, prevents isolation and allows the youngster to practice interactions in egalitarian relationships and to develop skills to cope with the sameness and the difference between their values, beliefs, attitudes and behaviours and those of others. Additionally, the peer group represents the system wherein adolescents will create intimate friendships and experience their first romantic/sexual relationships (Tuval-Mashiach et al., 2008).
Attachment to school has also been described as a strong correlate of well-being and successful development, as it contributes to the achievement of life goals, the learning and practice of social norms, trust in social relationships, a sense of competence and, consequently, higher self-esteem. All of these variables are known to prevent engagement in risk or maladaptive trajectories. Nevertheless, the literature on contextual and social support has emphasized its fundamental role, preventing adolescents from engaging in risk behaviours including sexual risk-behaviours, conduct disorders, alcohol abuse, delinquency and, above all, self-destructiveness (Deliberto & Nock, 2011;Logan-Greene et al., 2011).

Risky and maladaptive trajectories: from distress to self-destructiveness
Adolescent vulnerability to negative life events and subsequent distress does not arise only from the events per se but is also inf luenced by adolescents' perceptions of their own resources for coping with the stressors (Lazarus & Folkman, 1984). When adolescents perceive a lack of resources for coping with a specific situation, they will perceive it as being a 'crisis'. This means that 'deviations' to positive trajectories may occur as the youngsters engage in inadequate or negative coping strategies (Weyers, Ising, & Janke, 2005), launching them into risk trajectories or even into maladaptive trajectories which, at worst, may include the manifestation of psychopathology, and, in particular, self-destructiveness (Meehan, Peirson, & Fridjhon, 2007). Given the availability of resources and the adolescent's perception of such resources to prevent or trigger a crisis, the literature has emphasized individual, familial and other potential contextual protective and risk factors, which are present together in a constant and dynamic interaction (Van Orden et al., 2010).
In addition to family factors, previous studies have emphasized negative individual factors, including self-concept, low self-esteem, hopelessness, a perception of lack of self-control, ineffective coping strategies, and rigid cognitive styles, among issues, as being strongly associated with distress and risk trajectories (Loh & Wragg, 2004). The literature has also pointed to hopelessness, negativism, excessive consumption of alcohol and drugs, and psychological symptoms, namely, depressive symptoms (Fidan, Ceyhun, & Kirpinar, 2011;Ougrin et al., 2012) as additional risk factors specifically for maladaptive trajectories and selfdestructiveness. These self-destructive trajectories seem to show differences according to sex and age. Females tend to show more self-destructive behaviours than males, although males manifest more severe thoughts and behaviours. Moreover, older adolescents manifest more severe and intentional thoughts and behaviours, while younger adolescents manifest more self-harm behaviours without the intent of suicide (Ougrin et al., 2012).
Several family factors -e.g., poor family functioning, miscommunication between family elements, lack of family cohesion, low family warmth and emotional support, low monitoring, parental hostility, rejection, and family structural configuration (non-nuclear families) -seem to be correlated with self-destructiveness (Cruz et al., 2013;Jurich, 2008;Randell et al., 2006;Van Renen & Wild, 2008). Moreover, insecure attachment to caregivers plays a significant role in this issue, as it is related to adolescents' negative self-representations and can lead to low self-esteem, hopelessness, and difficulties in regulating emotions, all of which can prompt SDTB (Fidan et al., 2011).
A host of diverse risk factors from other systems may also be relevant to risky and maladaptive trajectories, namely, a lack of school attachment and low academic performance, integration into risky peer groups, experiences of victimization, breakups or unsuccessful romantic and sexual experiences (Tuval-Mashiach et al., 2008), as well as more distal factors including a lack of financial resources to deal with hedonistic values and social pressures towards consumerism (Durán et al., 2011). These factors are associated particularly with the need for peer approval and partly with adolescents' hope for their future and their ability to define life goals and plans to achieve them (Cebulla, 2009).
In our study, we explored, through adolescents' own voices, their own 'maps' -their personal views about the factors associated with developmentally adaptive trajectories punctuated by well-being vs. risk trajectories characterized by distress or maladaptive trajectories characterized by self-destructiveness.

METHODOLOGY
Using a snow-ball sampling technique, we obtained a sample of convenience composed of 33 community adolescents aged between 13 and 21, all living in the greater Lisbon area. Participants were excluded for the presence of any psychological/psychiatric diagnosis, if they were attending psychotherapy or family therapy or if they had been attending any mental health treatment in the past. Fifteen adolescents were aged between 13-16 years (45.5%), and 18 were aged between 17-21 years (54.5%). Eighteen participants were males (54.5%), and 15 were females (45.5%). The sample was organized into five Focus Groups (FGs), each with six to eight adolescents. We grouped, boys and girls into agematched groups, composed of those in the younger and intermediate adolescent age range (13-16 years) as well as the older adolescent and emerging adulthood age range (17-21 years), in order to respect developmental differences (Jackson & Goossens, 2006).
The FGs were conducted in the Department of Psychology at the University of Lisbon. Each FG lasted approximately 90 minutes and was completely audio recorded. All participants were informed of the nature of the research, the aims of the study and the details of participation, including the study's interactive nature. Parents of minors were contacted to obtain their informed consent, and the participants themselves also gave informed consent. In accordance with ethical principles, all participants were offered, if and when they felt it necessary, free clinical services.
The FG method was chosen because it has two primary advantages: the richness of the data that arises from the circular information exchange during interaction and dialogue, which gives depth and breadth to the themes explored; and the phenomenon of being in a group, which decreases the youngsters' sense of selfexposure when discussing sensitive themes (Bagnoli & Clark, 2010).
An interview guide with open-ended questions, following norms for the construction of a focus group from a published interview guide (Bagnoli & Clark, 2010), were used to generate discussion about factors that influence adolescents' well-being, distress and SDTB. The researcher acted as a catalyst, clarifying and mediating the role of the participants. At the end of the FG, a questionnaire on socio-demographic information was completed by all participants.
All FG session audio records were transcribed and reviewed by three psychologists from the research team for accuracy. Qualitative content analysis was performed using the qualitative data analysis software NVIVO 8. The content analysis was coded in three phases as follows: descriptive coding ('Who are the participants?' i.e., participants' attributes), thematic coding ('What are the participants talking about?' i.e., the key themes were identified mainly based on the nodal interview questions: well-being, distress and SDTB), and analytic coding from a bottom-up perspective and with constant comparisons among codes and transcriptions. These methods contributed to the constant development of a more ref lective and profound analysis ('What does this mean? Which ideas are emerging? How are these ideas aggregated or related?' i.e., ref lexion and interpretation process about meanings and relations between concepts or ideas). Although the analytic process of coding and interpretation was completed mainly by one researcher, another team researcher reviewed the scheme, and a third independent researcher was consulted whenever doubts persisted, to strengthen the validity of the findings (Richards, 2009).

FINDINGS 'What factors may contribute to the well-being of adolescents?'
Family, Individual and Friend/Peer factors emerged as the most relevant factors in adolescents' well-being. Family was the category most often cited by the majority of participants (Figure 1).

Family factors
Family emerged as the most referenced category by both males and females. The data revealed that guidance, emotional support, and the family relational climate and dialogue are the main family factors that contribute to well-being ( Figure 1). These adolescents' views are consistent with the literature which points to family as an important and secure 'backup' for adolescents (Soares, 2000). Interestingly, although adolescence is frequently regarded as a phase of struggle for independence and autonomy (Kocayörük, 2010), guidance was associated with behavioural regulation and control for our participants, suggesting that parental guidance is necessary to successfully cope with the typical challenges of adolescence and even with the autonomy process. As one participant stated, "We must have some kind of control, someone who guides us…" (Male [M], 21).
Emotional support emerged as an expression of positive affection, support, care and warmth. Also noteworthy is the finding that the family relational climate, which refers to the quality of family functioning and relationships, clearly plays a powerful role in well-being, as this category was cited more frequently, and by more participants, even than positive affection and support. These adolescents stated that family, as a whole, is a pivotal part of the family system (Kalafat, 2005).
According to the participants, even other subsystems (e.g., marital; fraternal) have an impact on adolescents' well-being: "It is not just the relationship of adolescents with their parents, but also between the parents themselves, because it is important that adolescents grow in a context where they feel a good relationship between them…" (M., 16).
The emerging data suggest that family cohesion is critical as a protective factor for adaptive adolescent trajectories. This finding is consistent with Reinherz and colleagues (2008), who demonstrated that family cohesion is correlated with adolescents' psychological well-being, to improved school performance and to decreased risk for maladaptive trajectories. Family dialogue also emerged as an important category, as adolescents considered this factor to be a source of security and support that improves their well-being: "In this stage, it is super relevant to talk with parents, to have an open communication and availability to… for them to listen to us" (Female [F.], 21).
Girls appear to emphasize family factors more than boys, which could be explained by the fact that females are more socially oriented with regard to relationships (Perrin et al., 2011). Younger adolescents referenced family factors more often than did older adolescents, which may signify that the former are in a more intermediate stage of autonomy and, consequently, are more dependent on family guidance and support.

Individual factors
Through our participants' reports, we inferred that individual factors are also essential protective factors for well-being, especially for socio-emotional strengths and freedom-autonomy. Three major sub-categories of socio-emotional strengths, the most common individual factor, emerged: self-valuing, self-regulation and emotional balance (Figure 1). Self-valuing -self-esteem, self-confidence and attractive physical appearance -was the most cited socio-emotional factor and formed a foundation for adaptive trajectories. "If I don't feel good about myself, I can´t... I can´t belong to a peer group..." (F., 18). Indeed, the literature on self-esteem has highlighted this variable as being fundamental to adolescents' psychological adjustment (Pepi, Faria, & Alesi, 2006).
Self-regulation was associated with attaining life goals and with self-control, which included limits, negotiation, exploration, safety, needs, consciousness and decisions; "What I try is to think to myself 'I did this I won´t do it again because I know I've hurt someone' " (F., 18). Interestingly, it should be noted that regulation emerges both as a family and as an individual predictor of well-being. This finding is consistent with the idea expressed by adolescents that they need parental guidance to attain autonomy.
Emotional balance was also identified as meaningful for well-being, and this factor appeared frequently under the designation "middle-ground": "It takes a lot of willpower to achieve the middle-ground" (M., 17); "I think… maybe… I think that we only can reach the middle-ground with maturity" (M., 21).
Consistent with the literature, the second broad individual category emerging from adolescents' voices was freedom/autonomy, the attainment of which was noted to be a crucial task during adolescence (Jackson & Goossens, 2006): "To be free but with responsibility" (M., 18); "But we also have to win our freedom ... freedom is not just given." (M., 20). However, autonomy was not referenced alongside selfvalue. Does this omission mean that adolescents consider self-value to be a more important protective factor for well-being than freedom/autonomy? Should we assume that there can be no autonomy without self-value?
More males than females cited these types of factors. This finding is consistent with the more individualistic and instrumental orientation of males as compared to females (Perrin et al., 2011). In addition, older adolescents seemed to emphasize these factors more often than did younger adolescents, which could be due to the former's more consistent autonomy when compared with younger adolescents.

Peer factors
Concerning the relevance of friends for well-being in adolescence, two main topics emerged, cited equally by males and females and by older and younger participants: Integration and Support (Figure 1). "We can experience a negative family relationship but having a peer group may compensate for that" (M., 18). The voices of these adolescents seem to be consistent with the literature, which emphasizes that integration in peer groups prevents isolation and allows youngsters to develop social and intimacy skills (Jurich, 2008).

'What factors may contribute to adolescents' distress?'
As with well-being, three main categories predicting distress emerged: Family (most cited), Individuals, and Friends. Finances were more often cited and by more participants as a risk factor than as a factor in well-being, particularly by girls (Figure 2). This phenomenon could potentially be associated with social pressures towards consumerism, concerns with physical appearance, which could easily become a source of distress (Durán et al., 2011).

Family factors
The most cited distressing family factor was parental inhibition of freedom/ autonomy, which is in accordance both with the aforementioned data and the adolescents' own need for freedom/autonomy (Crespo et al., 2011;Van Renen & Wild, 2008). Again, an apparent contradiction exists between parental inhibition of freedom/autonomy as a factor contributing to distress and parental regulation/ control as a factor contributing to well-being ( Figure 2). However, as has already been stated, this phenomenon could be a pseudo-contradiction, as adolescents' seem to demand progressive guidance from parents, which in fact grants them more autonomy. Addressing this issue, an adolescent said: "Someone that has no freedom inside the family (...) one day, when alone, when independent from the parents, that person may start acting out and behaving in disproportionate ways that may lead to bad paths" (F., 18).
A lack of emotional support, meaning that parents are not always responsive to adolescents' needs for support, was the second most cited factor, as expected (Van Orden et al., 2010). Although this is a family centrifugal phase, adolescents still perceive family as being the primary source of support to which they can return whenever they need.
We must highlight the emergence of parental hostility as a subcategory, which was referenced by almost one third of the participants. Parental authoritarianism or even family violence is far beyond even unresponsiveness or lack of emotional support. As we know, the literature also points to parental hostility as being a strong risk factor in adolescence (Cheng et al., 2009;Van Orden et al., 2010).
Consistent with the literature, a negative family climate was associated with adolescent distress just as a positive family climate was associated with well-being, thus reinforcing the role of the relational whole as a potential protective or risk factor for well-being (Jurich, 2008). A participant expressed: "Undoubtedly, conflicts at home! Because I think that affects the mental health of children, adolescents, and everyone..." (F., 21).
Family factors were similarly referenced by boys and girls, suggesting a consensus about the strong influence of family distress. Younger adolescents referenced family factors more often than did older adolescents, which may signify, as we have already hypothesized, that the former are more dependent on family guidance and support than the latter.

Individual factors
Two main categories emerged concerning individual factors: unmet needs and socio-emotional vulnerabilities (Figure 2). Adolescents pointed mainly to the need to be approved/accepted and the need to achieve established goals. The lack of acceptance by others has been considered as a risk factor for well-being, being associated with increased risk behaviours, isolation, decreased self-esteem and symptoms of psychopathology (Heilbron & Prinstein, 2010): "If we are marginalized... It counts a lot and our self-esteem decreases so much!" (M., 15).
Adolescence involves responding to multiple challenges, adapting to multiple changes, overcoming limits, making choices and making decisions, namely about school or work, all of which are necessary to establish and reach goals which seem to be essential to the youngsters' present and future lives (Loh & Wragg, 2004). When they cannot achieve their aims, adolescents experience failure and feel powerless and distressed. As mentioned by an adolescent: "as bigger they are [goals]and as more important we consider them, if we can't make it... obviously it would be bad for... for ourselves... we would feel badly..." (F., 18).
Our participants emphasized negative personal characteristics, values, and a lack of the skills that are considered to be necessary for well-being as socio-emotional vulnerabilities: "…it shows lack of autonomy... lack of character..." (M., 21). Self-devaluation, particularly self-esteem, was also highlighted as a point of socio-emotional vulnerability. Therefore, self-(de)valuing seems to be very relevant both as a protective and as a risk factor for adaptive trajectories in adolescence, which is consistent with the literature (Loh & Wragg, 2004;Pepi et al., 2006).
Again, boys referenced individual factors more often than did girls (Perrin et al., 2011). Younger and older adolescents showed a similar focus on individual factors.

Peer factors
Integration difficulties with peer groups and separation were the categories most referenced by the most participants ( Figure 2), with similar relevance for both sexes and age groups. This finding is also consistent with the literature that emphasizes how peer groups -or the lack of them -impact on adolescents' well-being and distress (Heilbron & Prinstein, 2010). According to our data, the (negative) quality of peer relationships appears not to be as relevant a factor as the integration with the group, which drives us to three ref lections. Does 'not belonging' to a peer group contribute more to distress than the negative characteristics of the group members or the negative quality of the relationships? Is it possible that in risk trajectories, there is an inf lated need for belonging to a group, whatever that group may be? Could an insufficient appreciation of relationship quality within peer groups reinforce risky trajectories? These questions should be a target for ref lection for researchers, caregivers, school staff and mental health technicians. Preventive interventions should be focused on the importance of the peer group and its relational quality for adolescents' well-being.

'What factors may contribute to adolescents' self-destructive thoughts and behaviours?'
Participants' 'voices' revealed individual factors as being the most prominent category, clearly far away from the relevance of others, even from family factors and romantic factors, which were also highly referenced (Figure 3). This phenomenon may signify, according to the adolescents, a greater responsibility of family on well-being and distress, and, consequently, on adaptive or risk trajectories as well as a greater responsibility of individuals on maladaptive trajectories with SDTB.

Individual factors
A main category emerged from the data on individual factors: negative or inadequate coping strategies. As Lazarus and Folkman (1984) suggested, stress is mainly induced by the evaluation of the stimulus based upon the self-perception of resources and skills. Negative or inadequate coping strategies, which represent ways of dealing with adversity that might augment stress (Weyers et al., 2005), and has been strongly associated with adolescent self-destructiveness (Meehan et al., 2007). The coping inferred from our data was (Figure 3): looking for positive feelings via SDTB (e.g., self-relief or relief from others, not feeling or ending psychological suffering), punishing/blaming others or self-punishment/ blame, avoiding problems or demanding help. Some examples of inadequate coping included: "to make parents feel the pain of loss" (M., 21); "maybe to forget [suffering]" (M., 16).
Approximately one third of the participants who discussed factors that contribute to adolescents' SDTB referenced personal characteristics and skills, negative feelings and self-devaluation, mainly, low self-esteem, which is consistent with the literature (Figure 3; Loh & Wragg, 2004): "They feel desperate" (M., 21); "Exactly!, they don´t like themselves and think they should be punished!" (M., 17).
In accordance with Perrin et al. (2011), boys referenced individual factors more often than girls did. Younger adolescents referred more to individual factors than older adolescents did. It is possible that, as younger adolescents are less independent from family and less autonomous, they feel pushed to "defend" family and attribute the responsibility for self-destructiveness more to the adolescent than to the family.

Family factors
Concerning family factors, the most frequently mentioned category, referenced by all participants, was negative family climate. Our data suggest that family climate seems to be even more relevant to self-destructive trajectories than to risky trajectories or to adaptive ones because (1) this factor was referenced by all participants, who believed that family factors contributed to destructive trajectories; and (2) more than half of the references to family factors, in this theme, were about negative family climate. This finding, again, seems to reinforce the power of the family as a whole (e.g., relational dynamic, affectivity, dialogue, cohesion, adaptability) to inf luence maladaptive trajectories and, specifically, adolescent self-destructiveness (Reinherz et al., 2008). Approximately half of our participants also referred to low emotional support, in accordance with the literature (Deliberto & Nock, 2011;Figure 3). As examples: "Family relationships are…because it is our support, our base, and if we are bad with our base, we are bad with all" (M., 16).
Family factors were similarly referenced by both males and females, suggesting, once again, that adolescents are consensual about their view of family as being pivotal in self-destructiveness. Older adolescents referred more often to family factors more than younger adolescents did, thus reinforcing the idea that the former attribute more responsibility to family while the latter place the responsibility mainly on their own shoulders.

Romantic factors
Peer factors were more relevant in our sample than romantic factors with regard to well-being and distress, though an inverse relationship was present for self-destructiveness. Together with a lack of family support or with low self-esteem, romantic factors seem to gain additional power, influencing SDTB, namely, conflicts, dependences and break-ups: "If I think I'm a mess and the only support I have is my boyfriend, if he suddenly break-up with me... I'm really a mess..." (F., 18).
Romantic factors were almost exclusively referenced by girls, and mainly by older girls, which may reflect that females are more focused on relationships and their affects (Perrin et al., 2011) or that both girls and older adolescents are more aware of and engaged in romantic and intimacy issues than younger adolescents.

Adolescents' maps of their own trajectories: a holistic analysis
It was our aim to explore, through adolescents' own voices, their own 'maps' of the factors that contribute to developmentally adaptive trajectories characterized by well-being, risk trajectories dominated by distress, and trajectories of SDTB.
Our holistic analysis revealed that: (1) girls talked more than boys about wellbeing and about SDTB; (2) girls talked as much about well-being as about distress; (3) boys talked more about distress than about well-being; (4) younger adolescents talked more than older adolescents about well-being; (5) younger adolescents talked as much about well-being as about distress; (6) older adolescents talked more about distress than about well-being; (7) participants talked much less about self-destructive thoughts and behaviours than about distress and well-being.
It is understandable that participants in the younger and intermediate stages of adolescence would reflect more easily on adaptive trajectories than those in the later adolescence stages would, because the former are still very much overwhelmed by an 'in-satisfied-family' view, whereas the latter, more autonomous, adolescents have more experience and are standing on the brink of adulthood. Older adolescents may already be constrained by an 'out-satisfied-family' view and be more focused on the negative aspects of adolescence. A similar explanation could be given regarding the differences between girls and boys: it is possible that girls tend to have a more 'in-satisfied-family' view, while boys have much more of an 'out-of-satisfied-family' view (Perrin et al., 2011).
Adolescents were less interventive, about SDTB, suggesting that these types of trajectories are more difficult topics for youngsters to discuss. Moreover, we wonder whether adolescents have a need and the opportunity to talk about this issue, or whether the social stigma inherent in discussions of death and, particularly, of suicide and self-harm behaviours, decreases adolescents' opportunities to reflect upon, question and become informed about self-destructiveness (Curtis, 2010).
All of these issues -gender and age differences, and the silence about selfdestructiveness -should be reflected upon the prevention of risk trajectories as well as for therapeutic interventions with adolescents.
Analysing the commentary from our participants, we developed a comprehensive and holistic map (Figure 4) of the main factors that contribute to each of the three trajectories already mentioned -adaptive, risky, and maladaptive. Looking through this map, we conclude that adolescents highlight Family Relational Climate, Emotional Support and Self-Esteem, as these factors correspond to emergent categories that are common to all the trajectories. The relevance of these factors to adolescents' trajectories is well documented in the literature (Reinherz et al., 2008). Comparing an adaptive trajectory, characterized by well-being, with a risk trajectory of distress, we find that, in addition to the three categories mentioned above, these two trajectories also have in common Family Regulation, Self-Regulation, and the peer factors designated by Integration and Apartness. Comparing the risk trajectory with the maladaptive trajectory, we observe that, in addition to Family Relational Climate, Emotional Support and Self-Esteem, these trajectories also have in common Personal Characteristics and Skills.
Adaptive trajectories specifically emphasize Family Dialogue, Emotional Balance and Freedom. Risk trajectories are specifically associated with Parental Inhibition of Freedom/Autonomy, Parental Hostility, Unmet Needs and Finances. Finally, SDTB trajectories are specifically associated with Negative Coping and Negative Romantic Factors.
As the literature on adolescence emphasizes (Kocayörük, 2010), Freedom/ Autonomy is an important individual factor that contributes to well-being and, by the same token, Parental Inhibition of Freedom/Autonomy is a factor which contributes to distress. As we have already discussed, the simultaneous relevance attributed to Parental Inhibition of Freedom/Autonomy as a factor contributing to distress, and to Parental Regulation as a factor contributing to well-being, could represent a pseudo-contradiction, with adolescents demanding from parents regulation or control while still moving towards increased autonomy.
Our maps of adolescent trajectories derived from focus group data prompted us to focus our hypotheses on individual and family factors and to explain how the interactivity between family and individual factors may foster adaptive, risky or selfdestructive trajectories. Family relational climate and parental emotional support create a context where 'self-regulated autonomy' is possible and well accepted by the adolescent. This parental regulation fosters positive self-valuation and, particularly, self-esteem. In a circular fashion, an adolescent with 'self-regulated autonomy' and high self-esteem reinforces the family relational climate, meaning that a selfmaintenance cycle is established that favours an adaptive trajectory. In the absence of this systemic configuration of family and individual factors, specifically with a negative family relational climate, lack of emotional support, inadequate parental regulation, weak self-regulation and weak self-esteem, and, consequently, negative strategies of coping, the adolescent incurs a risk trajectory. A long continuation of these trajectories, including the maintenance or escalation of the systemic configuration of the above factors and perhaps others, may increment psychological suffering, thereby favouring maladaptive trajectories which, at worst, may include SDTB.

Final reflections, limitations and implications
The individual voices of adolescents allow us to understand their own personal maps of the primary factors that contribute to adaptive trajectories characterized by well-being, risky trajectories overwhelmed by distress and maladaptive trajectories characterized by self-destructive thoughts and behaviours. These maps, which are, in general, consistent with the literature, lead us to a systemic hypothesis focused on individual and family factors, to explain how the interactivity between family and individual factors may foster adaptive, risk or self-destructive trajectories. However, these maps indicated the need for further research to deeply explore this emerging hypothesis.
This study has several limitations, namely, the exclusive focus on adolescents' commentary, the snow-ball recruitment sampling technique, the control of the sample inclusion criteria only based upon the information given by the adolescents or their families, and the subjectivity inherent in the process of qualitative data analysis. To strengthen the validity of the findings, all of these limitations should be considered in future studies.
Notwithstanding these limitations, this study can contribute to a better understanding of these trajectories and can enrich reflections about the conceptualization and implementation of prevention programs and therapeutic interventions. Specifically, this study can inform: (1) specific factors and processes that adolescents consider to be pivotal in order to increase adaptive trajectories or, conversely, to increase the risk of maladaptive trajectories; (2) differences in meanings and views according to sex and age; (3) specific difficulties in discussing adolescent self-destructiveness; and (4) the integration of the entire family system in preventive and clinical contexts; and (5) the inflated need for belonging to peer groups, with a possible minor consideration of its characteristics and relational qualities.
Qualitative research on adolescence seems to be very relevant, especially studies that consider the value of learning about adolescents, parents, teachers and technicians' views as experts of their own experience (Mitchell et al., 2010). The present study also intends to inspire quantitative research, namely, longitudinal studies, as it allows for the study of patterns and changes, using large samples, and, even, the generalization of results.